Percentage Measures - differentiate between actual zero and zero records

[Martyn Houghton]

At the moment when you undertake percentage measure, it is not able to differentiate between actual zero (i.e. you failed all the to meet the resolution SLA for all the records) and zero records (i.e. no requests for the measure were found).

Would it be possible for the measure to capture the difference between the actual zero value and the no records found, perhaps storing a Null value instead of a zero, then displaying it as a non-coloured '0' on the scorecard rather than a red one?

The other alternative is having the option to reverse the percentage calculation, in that your measure selection calculates the opposite, i.e. the percentage of records you failed the condition, then take the calculated value away from 100 (%) in the second phase of the calculation before storing the measure. value. That way when you have no failure percentage or no requests found the zero is taken away from 100% leaving 100%.

The latter has the advantage that you do not need to alter the storage or display processes for the measure, just an option flag and different calculation method.

Cheers

Martyn

[NeilWJ]

Hi Martyn,

apologies for the delay getting back to you on this. So can i just clarify what the problem is:-
 

Say you have a measure called Changes That Met SLA so something like:-

% of calls where SLAMET=1 and class='change'

If there are no actual records (there are no records of class 'change')  the stored sample is 0. (but in real terms there is nothing to worry about as there were no actual records)
If there were 50 change records and all of them had SLAMET=0 the stored sample is 0. (in real terms there is cause for concern)

So the problem is you can't tell if its 0% for real or 0% because there were no change records. The measure can then be misleading because your servicedesk manager thinks wow we have missed all our change slas when in fact you don't have any changes logged in the system.

Is that the issue?

Cheers

[Martyn Houghton]

@NeilWJ

That's spot on, in our case for Incidents, but the same thing applies.

Lots of Red Zero's. Dont't Panic!

Cheers

Martyn

[Martyn Houghton]

@NeilWJ

Just wondering if there was any update on this issue?

Cheers

Martyn

[NeilWJ]

@Martyn Houghton i will check with api team as i reported the issue to them. Will let you know.

Cheers

[NeilWJ]

@Martyn Houghton i have asked the server team for some feedback on fixing this. I did provide a solution to the problem when this was first raised so they can just can implement it (if they are happy with it). I will get a firm response next week. Cheers

[Martyn Houghton]

@NeilWJ

Thanks for following it up.

Cheers

Martyn

[Martyn Houghton]

@NeilWJ

Is there any update on this one from your discussion with the Server Team?

Cheers

Martyn

[NeilWJ]

Hi @Martyn Houghton,

I have just asked them for an update. Will let you know.

Cheers

[Martyn Houghton]

@NeilWJ

Thanks

Martyn